Two-photon microscopy, together with the use of genetically encoded fluorescent proteins (e.g. GFP), has made it possible to directly visualize cellular behavior in 3- dimensional living tissues. We have been using this approach to investigate thymocyte migration and interactions with thymic stromal cells, and how these events relate to positive and negative selection of the T cell antigen receptor (TCR) repertoire. In the previous funding period we made the unexpected observation that many thymocytes undergoing negative selection do not rapidly arrest and die, but rather adopt a highly confined migration pattern while make repeated, transient contacts with dendritic cells surrounding confinement zones. In Aim 1 we plan to 1) determine the anatomical features of confinement zones for auto-reactive medullary thymocytes, 2) test the hypothesis that a large proportion of polyclonal medullary thymocytes are undergoing confined migration as the result of negative selection, 3) explore the role of AIRE in establishing anatomical niches for negative selection and, 4) determine how the abundance of negative selecting ligands affects the behavior of auto-reactive thymocytes during negative selection. In Aim 2, we plan to investigate the relationship between migration, chemokines and TCR repertoire selection by 1) examining the balance between thymocyte migration and TCR signaling during different modes of thymocyte selection, 2) examining the impact of modulating chemokine gradients on thymocyte-stromal cell interactions and thymocyte selection, and 3) investigating the relationship between altered chemokine response and impaired positive selection in thymocytes lacking the adaptor protein, GIT-2. By providing insight into the mechanisms that generate mature T cells that are functional and self-tolerant, these studies may lead to improved treatments for autoimmunity, immunodeficiency, and improve our ability to manipulate immune response to tumors or pathogens.